Enable Link Time Optimizations #16791

pull elichai wants to merge 1 commits into bitcoin:master from elichai:2019-09-thinlto changing 2 files +28 −4
  1. elichai commented at 0:10 am on September 3, 2019: contributor

    Mostly an idea, hard to measure exact benefits, if anyone has any ideas of good benchmarks for this will be welcome, altough in theory this should increase performance.

    Minimal building + bench_bitcoin benchmark in the comment.

    EDIT: As fanquake pointed out, this was brought up before and there were some issues around IBD. I personally do not think that not stripped binary sizes really matter, because we care about sizes only when shipping to others(i.e. part of a linux dist) and then we strip anyway. I will conduct IBD tests. I have a feeling that this will show good results with clang(thin lto) and not so good ones with gcc. I’ll report results.

  2. Enable Link Time Optimizations 3643493b5d
  3. elichai commented at 0:10 am on September 3, 2019: contributor

    clang without LTO run 1

     0# Benchmark, evals, iterations, total, min, max, median
 1AssembleBlock, 5, 700, 1.75888, 0.000499424, 0.000509562, 0.000501284
 2Base58CheckEncode, 5, 320000, 4.44531, 2.76679e-06, 2.79008e-06, 2.78027e-06
 3Base58Decode, 5, 800000, 2.51903, 6.27043e-07, 6.33854e-07, 6.29572e-07
 4Base58Encode, 5, 470000, 4.544, 1.92153e-06, 1.9427e-06, 1.93532e-06
 5Bech32Decode, 5, 800000, 1.12439, 2.77588e-07, 2.83886e-07, 2.80391e-07
 6Bech32Encode, 5, 800000, 1.52587, 3.7683e-07, 3.84009e-07, 3.81521e-07
 7BenchLockedPool, 5, 1300, 3.97761, 0.000601612, 0.000639539, 0.000602839
 8BnBExhaustion, 5, 650, 2.60574, 0.000790225, 0.000820404, 0.000799679
 9CCheckQueueSpeedPrevectorJob, 5, 1400, 13.258, 0.00183946, 0.00196889, 0.00184942
10CCoinsCaching, 5, 170000, 0.328199, 3.58821e-07, 4.01443e-07, 3.91374e-07
11CoinSelection, 5, 650, 0.572936, 0.000171902, 0.000183726, 0.000176019
12ConstructGCSFilter, 5, 1000, 8.10747, 0.00153589, 0.00174128, 0.001621
13DeserializeAndCheckBlockTest, 5, 160, 4.65824, 0.00567065, 0.00629352, 0.00569696
14DeserializeBlockTest, 5, 130, 3.08922, 0.00465626, 0.00492937, 0.00473448
15DuplicateInputs, 5, 10, 0.357063, 0.00663135, 0.00819393, 0.00668833
16FastRandom_1bit, 5, 440000000, 3.66524, 1.66211e-09, 1.67058e-09, 1.66704e-09
17FastRandom_32bit, 5, 110000000, 4.597, 8.26189e-09, 8.46674e-09, 8.3029e-09
18MatchGCSFilter, 5, 50000, 8.19405, 3.22754e-05, 3.39363e-05, 3.25489e-05
19MempoolEviction, 5, 41000, 2.87917, 1.40002e-05, 1.40963e-05, 1.4055e-05
20MerkleRoot, 5, 800, 4.37644, 0.00109208, 0.00109532, 0.00109406
21PrevectorClearNontrivial, 5, 28300, 0.000286874, 1.66226e-09, 2.55887e-09, 1.75353e-09
22PrevectorClearTrivial, 5, 88600, 0.000678416, 1.51194e-09, 1.53922e-09, 1.53875e-09
23PrevectorDeserializeNontrivial, 5, 6800, 6.14197, 0.000179907, 0.000181122, 0.000180681
24PrevectorDeserializeTrivial, 5, 52000, 5.27403, 2.02025e-05, 2.03668e-05, 2.02626e-05
25PrevectorDestructorNontrivial, 5, 28800, 0.000242584, 1.57406e-09, 1.76795e-09, 1.69417e-09
26PrevectorDestructorTrivial, 5, 88900, 0.000692209, 1.41076e-09, 2.01678e-09, 1.45669e-09
27PrevectorResizeNontrivial, 5, 28900, 1.09466, 7.54659e-06, 7.63179e-06, 7.57301e-06
28PrevectorResizeTrivial, 5, 90300, 3.63849, 8.03433e-06, 8.08772e-06, 8.04891e-06
29RIPEMD160, 5, 440, 4.87181, 0.00220603, 0.00222362, 0.00221297
30RollingBloom, 5, 1500000, 3.624, 4.71539e-07, 4.92985e-07, 4.82655e-07
31SHA1, 5, 570, 4.6708, 0.00162474, 0.00166248, 0.00163211
32SHA256, 5, 340, 4.53006, 0.00262891, 0.00274689, 0.0026462
33SHA256D64_1024, 5, 7400, 4.87046, 0.000115724, 0.000142581, 0.000135341
34SHA256_32b, 5, 4700000, 4.58475, 1.91316e-07, 2.0184e-07, 1.94727e-07
35SHA512, 5, 330, 3.97634, 0.00235566, 0.00248732, 0.00239992
36SipHash_32b, 5, 40000000, 4.72604, 2.29608e-08, 2.44476e-08, 2.33909e-08
37Sleep100ms, 5, 10, 5.01552, 0.100257, 0.100413, 0.100312
38Trig, 5, 12000000, 0.761389, 1.1107e-08, 1.3417e-08, 1.3107e-08
39VerifyScriptBench, 5, 6300, 2.60303, 7.9582e-05, 8.53183e-05, 8.21487e-05

    clang without lto run 2

     0# Benchmark, evals, iterations, total, min, max, median
 1AssembleBlock, 5, 700, 1.96617, 0.000499029, 0.00061569, 0.000555522
 2Base58CheckEncode, 5, 320000, 4.58013, 2.85247e-06, 2.88703e-06, 2.85549e-06
 3Base58Decode, 5, 800000, 2.59868, 6.43124e-07, 6.6978e-07, 6.45136e-07
 4Base58Encode, 5, 470000, 4.73274, 1.99772e-06, 2.0328e-06, 2.01745e-06
 5Bech32Decode, 5, 800000, 1.17128, 2.91664e-07, 2.94089e-07, 2.92609e-07
 6Bech32Encode, 5, 800000, 1.58015, 3.93101e-07, 3.98773e-07, 3.94542e-07
 7BenchLockedPool, 5, 1300, 4.17851, 0.000625214, 0.000674724, 0.000633105
 8BnBExhaustion, 5, 650, 2.65366, 0.000803413, 0.000839476, 0.000811133
 9CCheckQueueSpeedPrevectorJob, 5, 1400, 13.386, 0.00183121, 0.00195832, 0.00193187
10CCoinsCaching, 5, 170000, 0.33303, 3.75554e-07, 4.06653e-07, 3.9148e-07
11CoinSelection, 5, 650, 0.549849, 0.0001686, 0.000170028, 0.000169009
12ConstructGCSFilter, 5, 1000, 7.81065, 0.00153585, 0.00160931, 0.00155086
13DeserializeAndCheckBlockTest, 5, 160, 4.3975, 0.00542555, 0.00555456, 0.00549932
14DeserializeBlockTest, 5, 130, 3.01934, 0.00452266, 0.00473618, 0.0046844
15DuplicateInputs, 5, 10, 0.341157, 0.00633856, 0.00768086, 0.00638198
16FastRandom_1bit, 5, 440000000, 3.63868, 1.61861e-09, 1.70596e-09, 1.64989e-09
17FastRandom_32bit, 5, 110000000, 4.5247, 7.9982e-09, 8.44111e-09, 8.28698e-09
18MatchGCSFilter, 5, 50000, 7.74736, 3.03944e-05, 3.26011e-05, 3.06948e-05
19MempoolEviction, 5, 41000, 2.98457, 1.41979e-05, 1.50077e-05, 1.44873e-05
20MerkleRoot, 5, 800, 5.06723, 0.00113876, 0.00143356, 0.00125824
21PrevectorClearNontrivial, 5, 28300, 0.000297168, 1.73883e-09, 2.80184e-09, 2.0053e-09
22PrevectorClearTrivial, 5, 88600, 0.000623501, 1.33983e-09, 1.52511e-09, 1.40331e-09
23PrevectorDeserializeNontrivial, 5, 6800, 6.34991, 0.000179312, 0.000194279, 0.000186073
24PrevectorDeserializeTrivial, 5, 52000, 5.48437, 2.05534e-05, 2.16369e-05, 2.10327e-05
25PrevectorDestructorNontrivial, 5, 28800, 0.000243668, 1.5741e-09, 2.10503e-09, 1.57552e-09
26PrevectorDestructorTrivial, 5, 88900, 0.000728, 1.47732e-09, 2.03974e-09, 1.5748e-09
27PrevectorResizeNontrivial, 5, 28900, 1.12402, 7.61811e-06, 8.15651e-06, 7.66037e-06
28PrevectorResizeTrivial, 5, 90300, 3.74789, 7.93443e-06, 8.72624e-06, 8.28083e-06
29RIPEMD160, 5, 440, 5.06353, 0.00226185, 0.00238738, 0.00226887
30RollingBloom, 5, 1500000, 4.37212, 4.89467e-07, 7.16985e-07, 5.49323e-07
31SHA1, 5, 570, 4.97618, 0.00164691, 0.0018421, 0.00175501
32SHA256, 5, 340, 4.76886, 0.002788, 0.00284018, 0.00279755
33SHA256D64_1024, 5, 7400, 4.36494, 0.000117053, 0.000119305, 0.000117821
34SHA256_32b, 5, 4700000, 4.78085, 2.01433e-07, 2.09491e-07, 2.0206e-07
35SHA512, 5, 330, 4.02069, 0.00239671, 0.00248702, 0.00243321
36SipHash_32b, 5, 40000000, 4.7499, 2.34689e-08, 2.42086e-08, 2.35923e-08
37Sleep100ms, 5, 10, 5.01435, 0.100229, 0.100332, 0.100288
38Trig, 5, 12000000, 0.673552, 1.0744e-08, 1.22859e-08, 1.10018e-08
39VerifyScriptBench, 5, 6300, 2.60643, 8.2142e-05, 8.32055e-05, 8.27119e-05

    clang with lto run 1

     0# Benchmark, evals, iterations, total, min, max, median
 1AssembleBlock, 5, 700, 1.72164, 0.000488554, 0.000495535, 0.00049131
 2Base58CheckEncode, 5, 320000, 4.15869, 2.59372e-06, 2.60309e-06, 2.60095e-06
 3Base58Decode, 5, 800000, 2.17661, 5.41487e-07, 5.4834e-07, 5.44013e-07
 4Base58Encode, 5, 470000, 4.14466, 1.75899e-06, 1.76838e-06, 1.76369e-06
 5Bech32Decode, 5, 800000, 1.19821, 2.97339e-07, 3.00996e-07, 2.99391e-07
 6Bech32Encode, 5, 800000, 1.62597, 4.03557e-07, 4.08587e-07, 4.0681e-07
 7BenchLockedPool, 5, 1300, 3.89975, 0.000597339, 0.000602308, 0.000599304
 8BnBExhaustion, 5, 650, 2.36543, 0.00072606, 0.000729458, 0.00072771
 9CCheckQueueSpeedPrevectorJob, 5, 1400, 12.4383, 0.00173862, 0.0018336, 0.00177781
10CCoinsCaching, 5, 170000, 0.312746, 3.501e-07, 3.79826e-07, 3.68905e-07
11CoinSelection, 5, 650, 0.546785, 0.000166795, 0.000169946, 0.000168376
12ConstructGCSFilter, 5, 1000, 7.03575, 0.00140299, 0.0014086, 0.00140827
13DeserializeAndCheckBlockTest, 5, 160, 4.18461, 0.00522168, 0.00523717, 0.00523443
14DeserializeBlockTest, 5, 130, 2.87828, 0.00441814, 0.00444233, 0.00442485
15DuplicateInputs, 5, 10, 0.328547, 0.00644949, 0.00681943, 0.00654519
16FastRandom_1bit, 5, 440000000, 3.2298, 1.46282e-09, 1.4778e-09, 1.46598e-09
17FastRandom_32bit, 5, 110000000, 4.31731, 7.82775e-09, 7.86024e-09, 7.85525e-09
18MatchGCSFilter, 5, 50000, 6.96338, 2.77827e-05, 2.78858e-05, 2.7871e-05
19MempoolEviction, 5, 41000, 2.47832, 1.20621e-05, 1.21223e-05, 1.20881e-05
20MerkleRoot, 5, 800, 4.29475, 0.00106896, 0.00107777, 0.00107411
21PrevectorClearNontrivial, 5, 28300, 0.000235125, 1.41194e-09, 1.75353e-09, 1.72703e-09
22PrevectorClearTrivial, 5, 88600, 0.000636042, 1.38168e-09, 1.45128e-09, 1.44847e-09
23PrevectorDeserializeNontrivial, 5, 6800, 6.45151, 0.000188512, 0.00019084, 0.000190182
24PrevectorDeserializeTrivial, 5, 52000, 5.1855, 1.99341e-05, 1.99596e-05, 1.99413e-05
25PrevectorDestructorNontrivial, 5, 28800, 0.000230332, 1.57406e-09, 1.6334e-09, 1.58128e-09
26PrevectorDestructorTrivial, 5, 88900, 0.000636916, 1.39389e-09, 1.51434e-09, 1.42107e-09
27PrevectorResizeNontrivial, 5, 28900, 1.06484, 7.31545e-06, 7.40119e-06, 7.37174e-06
28PrevectorResizeTrivial, 5, 90300, 3.55419, 7.8563e-06, 7.88425e-06, 7.87187e-06
29RIPEMD160, 5, 440, 4.72668, 0.00214663, 0.00215083, 0.00214833
30RollingBloom, 5, 1500000, 3.433, 4.5134e-07, 4.60353e-07, 4.58777e-07
31SHA1, 5, 570, 4.62547, 0.00162106, 0.00162443, 0.00162312
32SHA256, 5, 340, 4.44984, 0.00260789, 0.00263764, 0.00261407
33SHA256D64_1024, 5, 7400, 4.23212, 0.000114082, 0.000114491, 0.000114457
34SHA256_32b, 5, 4700000, 4.30349, 1.81287e-07, 1.85391e-07, 1.81894e-07
35SHA512, 5, 330, 3.91958, 0.002371, 0.00238079, 0.00237517
36SipHash_32b, 5, 40000000, 4.53516, 2.26558e-08, 2.27179e-08, 2.26719e-08
37Sleep100ms, 5, 10, 5.02578, 0.100485, 0.100544, 0.100521
38Trig, 5, 12000000, 0.621274, 9.54652e-09, 1.06749e-08, 1.05556e-08
39VerifyScriptBench, 5, 6300, 2.58936, 8.19543e-05, 8.24921e-05, 8.2189e-05

    clang with lto run 2

     0# Benchmark, evals, iterations, total, min, max, median
 1AssembleBlock, 5, 700, 1.99427, 0.000493252, 0.000665168, 0.000559993
 2Base58CheckEncode, 5, 320000, 5.05568, 3.06908e-06, 3.25902e-06, 3.13718e-06
 3Base58Decode, 5, 800000, 2.95883, 6.62271e-07, 7.67973e-07, 7.58572e-07
 4Base58Encode, 5, 470000, 5.0744, 1.94787e-06, 2.43021e-06, 2.21603e-06
 5Bech32Decode, 5, 800000, 1.36161, 3.20821e-07, 3.8574e-07, 3.31155e-07
 6Bech32Encode, 5, 800000, 1.87327, 4.1505e-07, 5.15584e-07, 4.81936e-07
 7BenchLockedPool, 5, 1300, 4.97714, 0.000717744, 0.000807546, 0.000761353
 8BnBExhaustion, 5, 650, 2.9194, 0.000846583, 0.000942232, 0.000902514
 9CCheckQueueSpeedPrevectorJob, 5, 1400, 13.2917, 0.00177237, 0.00208845, 0.0018112
10CCoinsCaching, 5, 170000, 0.312361, 3.58964e-07, 3.74724e-07, 3.69679e-07
11CoinSelection, 5, 650, 0.544713, 0.000166541, 0.000169268, 0.000167463
12ConstructGCSFilter, 5, 1000, 7.76531, 0.00150814, 0.0016057, 0.0015303
13DeserializeAndCheckBlockTest, 5, 160, 4.32749, 0.00533709, 0.00547389, 0.00540589
14DeserializeBlockTest, 5, 130, 2.9427, 0.00449615, 0.00455946, 0.00452504
15DuplicateInputs, 5, 10, 0.31676, 0.0063014, 0.00635066, 0.0063463
16FastRandom_1bit, 5, 440000000, 3.50853, 1.58325e-09, 1.60725e-09, 1.59132e-09
17FastRandom_32bit, 5, 110000000, 4.42874, 7.95085e-09, 8.17166e-09, 8.05469e-09
18MatchGCSFilter, 5, 50000, 8.16707, 3.12369e-05, 3.47419e-05, 3.22018e-05
19MempoolEviction, 5, 41000, 3.01924, 1.31778e-05, 1.61007e-05, 1.46303e-05
20MerkleRoot, 5, 800, 4.27195, 0.00103943, 0.00113243, 0.00106114
21PrevectorClearNontrivial, 5, 28300, 0.000258918, 1.49e-09, 2.61336e-09, 1.69908e-09
22PrevectorClearTrivial, 5, 88600, 0.000699667, 1.44422e-09, 1.75696e-09, 1.53358e-09
23PrevectorDeserializeNontrivial, 5, 6800, 5.88229, 0.000172483, 0.000173993, 0.000172633
24PrevectorDeserializeTrivial, 5, 52000, 5.05375, 1.94089e-05, 1.94787e-05, 1.94224e-05
25PrevectorDestructorNontrivial, 5, 28800, 0.000259292, 1.57552e-09, 2.55642e-09, 1.63194e-09
26PrevectorDestructorTrivial, 5, 88900, 0.00065175, 1.39061e-09, 1.57434e-09, 1.44263e-09
27PrevectorResizeNontrivial, 5, 28900, 1.03944, 7.14281e-06, 7.25415e-06, 7.20041e-06
28PrevectorResizeTrivial, 5, 90300, 3.45045, 7.62638e-06, 7.65751e-06, 7.63868e-06
29RIPEMD160, 5, 440, 4.60618, 0.00208886, 0.00210049, 0.00209236
30RollingBloom, 5, 1500000, 3.35536, 4.46947e-07, 4.47789e-07, 4.47367e-07
31SHA1, 5, 570, 4.50482, 0.00157792, 0.00158378, 0.00157961
32SHA256, 5, 340, 4.30758, 0.00250335, 0.00257137, 0.00252218
33SHA256D64_1024, 5, 7400, 4.04927, 0.000108863, 0.00011012, 0.000109247
34SHA256_32b, 5, 4700000, 4.50374, 1.89222e-07, 1.94606e-07, 1.90612e-07
35SHA512, 5, 330, 3.90106, 0.00230413, 0.00246755, 0.00234779
36SipHash_32b, 5, 40000000, 4.54047, 2.24717e-08, 2.28638e-08, 2.27357e-08
37Sleep100ms, 5, 10, 5.0232, 0.100405, 0.100544, 0.10044
38Trig, 5, 12000000, 0.607282, 9.90421e-09, 1.03326e-08, 1.01497e-08
39VerifyScriptBench, 5, 6300, 2.45329, 7.70539e-05, 7.95727e-05, 7.75427e-05
  4. fanquake added the label Build system on Sep 3, 2019
  5. elichai commented at 0:12 am on September 3, 2019: contributor

    Build time benchmark (with ./configure --with-incompatible-bdb)

    1. Clang without LTO:
    0real    3m59.577s
1user    54m18.507s
2sys     2m7.528s
    1. Clang with LTO:
    0real    7m7.265s
1user    97m1.044s
2sys     2m49.948s
    1. GCC without LTO:
    0real    3m13.461s
1user    37m23.436s
2sys     2m8.695s
    1. GCC with LTO:
    0real    6m20.885s
1user    40m0.574s
2sys     2m53.457s
  6. fanquake commented at 0:14 am on September 3, 2019: member
    Have you read through the past (#10616, #10800) and current (#14277) discussions around enabling LTO? If not, that will likely give you a starting point for performance measurement, build system considerations etc.
  7. elichai commented at 0:18 am on September 3, 2019: contributor
    @fanquake I knew I forgot something hehe, i’ll go read them now. thanks!
  8. DrahtBot commented at 3:10 am on September 3, 2019: member

    The following sections might be updated with supplementary metadata relevant to reviewers and maintainers.

    Conflicts

    Reviewers, this pull request conflicts with the following ones:

    • #16834 (Fetch Headers over DNS by TheBlueMatt)
    • #16762 (Rust-based Backup over-REST block downloader by TheBlueMatt)

    If you consider this pull request important, please also help to review the conflicting pull requests. Ideally, start with the one that should be merged first.

  9. practicalswift commented at 12:07 pm on September 3, 2019: contributor

    Concept ACK assuming the default is switched: LTO should be opt-in using --enable-lto to allow for risk-free experimentation and for the reason @laanwj gives in #10616 (comment):

    It should definitely not be enabled by default! Programs usually shouldn’t add non-standard compilation flags by default unless necessary.

  10. laanwj commented at 12:54 pm on September 3, 2019: member

    I think this is an interesting experiment!

    However, build-system-wise, this simply adds some compiler and linker flags, which could be passed in through CFLAGS, CXXFLAGS, CPPFLAGS, LDFLAGS environment variables. I don’t think it really belongs as a separate configure option for individual applications.

    Will leave it to @theuni though.

  11. laanwj assigned theuni on Sep 3, 2019
  12. laanwj commented at 8:45 am on October 2, 2019: member
    This is unlikely to be merged. Closing this PR. (feel free to continue discussion about LTO in the release builds, of course)
  13. laanwj closed this on Oct 2, 2019
  14. fanquake referenced this in commit 681b25e3cd on Nov 25, 2021
  15. sidhujag referenced this in commit 67658eec7d on Nov 25, 2021
  16. DrahtBot locked this on Dec 16, 2021


elichai fanquake DrahtBot practicalswift laanwj

Labels
Build system

github-metadata-mirror

This is a metadata mirror of the GitHub repository bitcoin/bitcoin. This site is not affiliated with GitHub. Content is generated from a GitHub metadata backup.
generated: 2024-09-29 01:12 UTC
This site is hosted by @0xB10C
More mirrored repositories can be found on mirror.b10c.me